Development and validation of machine learning model for predicting treatment responders in patients with primary biliary cholangitis.

AIMS: Ursodeoxycholic acid is the first-line treatment for primary biliary cholangitis, and treatment response is one of the factors predicting the outcome. To prescribe alternative therapies, clinicians might need additional information before deciphering the treatment response to ursodeoxycholic acid, contributing to a better patient prognosis. In this study, we developed and validated machine learning (ML) algorithms to predict treatment responses using pretreatment data. METHODS: This multicenter cohort study included collecting datasets from two data samples. Data 1 included 245 patients from 18 hospitals for ML development, and was divided into (i) training and (ii) development sets. Data 2 (iii: test set) included 51 patients from our hospital for validation. An extreme gradient boosted tree predicted the treatment response in the ML model. The area under the curve was used to evaluate the efficacy of the algorithm. RESULTS: Data 1 showed that patients complying with the Paris II treatment response had significantly lower serum alkaline phosphatase and total bilirubin levels than those who did not respond. Three factors, total bilirubin, total protein, and alanine aminotransferase levels were selected as essential variables for prediction. Data 2 showed that patients complying with the Paris II criteria had significantly high prothrombin time and low total bilirubin levels. The area under the curve of extreme gradient boosted tree was good for (ii) (0.811) and (iii) (0.856). CONCLUSIONS: We demonstrated the efficacy of ML in predicting the treatment response for patients with primary biliary cholangitis. Early identification of cases requiring additional treatment with our novel ML model may improve prognosis.

Recent advances in hepatitis A virus research and clinical practice guidelines for hepatitis A virus infection in Japan.

In 2018, there was a hepatitis A outbreak in Japan, and hepatitis A virus (HAV) infection is considered a sexually transmitted disease. In general, patients with hepatitis A should be given attention, and this disease should be prevented more than ever. The Japan Agency for Medical Research and Development (AMED) Hepatitis A and E viruses (HAV and HEV) Study Group has worked on the project to create "Recent Advances in Hepatitis A Virus (HAV) Research and Clinical Practice Guidelines for HAV Infection in Japan". The group consists of expert hepatologists and virologists who gathered at virtual meeting on August 5, 2023. Data about the pathogenesis, infection routes, diagnosis, complications, several factors for the severities, vaccination, and current and future treatments for hepatitis A were discussed and debated for a draft version. The participants assessed the quality of cited studies. The finalized recommendations are presented in this review. The recent advances in HAV research and clinical practice for HAV infection in Japan, have been reviewed by the AMED HAV and HEV Study Group.

Analysis of distribution, collection, and confirmation of capacity dependency of small extracellular vesicles toward a therapy for liver cirrhosis.

BACKGROUND: The progression of liver fibrosis leads to portal hypertension and liver dysfunction. However, no antifibrotic agents have been approved for cirrhosis to date, making them an unmet medical need. Small extracellular vesicles (sEVs) of mesenchymal stem cells (MSCs) are among these candidate agents. In this study, we investigated the effects of sEVs of MSCs, analyzed their distribution in the liver post-administration, whether their effect was dose-dependent, and whether it was possible to collect a large number of sEVs. METHODS: sEVs expressing tdTomato were generated, and their uptake into constituent liver cells was observed in vitro, as well as their sites of uptake and cells in the liver using a mouse model of liver cirrhosis. The efficiency of sEV collection using tangential flow filtration (TFF) and changes in the therapeutic effects of sEVs in a volume-dependent manner were examined. RESULTS: The sEVs of MSCs accumulated mostly in macrophages in damaged areas of the liver. In addition, the therapeutic effect of sEVs was not necessarily dose-dependent, and it reached a plateau when the dosage exceeded a certain level. Furthermore, although ultracentrifugation was commonly used to collect sEVs for research purposes, we verified that TFF could be used for efficient sEV collection and that their effectiveness is not reduced. CONCLUSION: In this study, we identified some unknown aspects regarding the dynamics, collection, and capacity dependence of sEVs. Our results provide important fundamentals for the development of therapies using sEVs and hold potential implications for the therapeutic applications of sEV-based therapies for liver cirrhosis.

Machine learning prediction model for treatment responders in patients with primary biliary cholangitis.

Background and Aim: Treatment response to ursodeoxycholic acid may predict the prognosis of patients with primary biliary cholangitis (PBC). Recent studies have suggested the benefits of using machine learning (ML) to forecast complex medical predictions. We aimed to predict treatment response in patients with PBC using ML and pretreatment data. Methods: We conducted a single-center retrospective study and collected data from 194 patients with PBC who were followed up for at least 12 months after treatment initiation. Patient data were analyzed with five ML models, namely random forest, extreme gradient boosting (XGB), decision tree, naïve Bayes, or logistic regression, to predict treatment response using the Paris II criteria. The established models were assessed using an out-of-sample validation. The area under the curve (AUC) was used to evaluate the efficacy of each algorithm. Overall survival and liver-related deaths were analyzed using Kaplan-Meier analysis. Results: Compared to logistic regression (AUC = 0.595, P = 0.0219, 0.031 models), ML analyses showed significantly high AUC in the random forest (AUC = 0.84) and XGB (AUC = 0.83) models; however, the AUC was not significantly high for decision tree (AUC = 0.633) or naïve Bayes (AUC = 0.584) models. Kaplan-Meier analysis showed significantly improved prognoses in patients predicted to achieve the Paris II criteria by XGB (log-rank = 0.005 and 0.007). Conclusion: ML algorithms could improve treatment response prediction using pretreatment data, which could lead to better prognoses. In addition, the ML model using XGB could predict the prognosis of patients before treatment initiation.

Escherichia coli-derived outer-membrane vesicles induce immune activation and progression of cirrhosis in mice and humans.

BACKGROUND AND AIMS: Decompensated cirrhosis with fibrosis progression causes portal hypertension followed by an oedematous intestinal tract. These conditions weaken the barrier function against bacteria in the intestinal tract, a condition called leaky gut, resulting in invasion by bacteria and bacterial components. Here, we investigated the role of outer-membrane vesicles (OMVs) of Escherichia coli, which is the representative pathogenic gut-derived bacteria in patients with cirrhosis in the pathogenesis of cirrhosis. METHODS: We investigated the involvement of OMVs in humans using human serum and ascites samples and also investigated the involvement of OMVs from E. coli in mice using mouse liver-derived cells and a mouse cirrhosis model. RESULTS: In vitro, OMVs induced inflammatory responses to macrophages and neutrophils, including the upregulation of C-type lectin domain family 4 member E (Clec4e), and induced the suppression of albumin production in hepatocytes but had a relatively little direct effect on hepatic stellate cells. In a mouse cirrhosis model, administration of OMVs led to increased liver inflammation, especially affecting the activation of macrophages, worsening fibrosis and decreasing albumin production. Albumin administration weakened these inflammatory changes. In addition, multiple antibodies against bacterial components were increased with a progressing Child-Pugh grade, and OMVs were detected in ascites of patients with decompensated cirrhosis. CONCLUSIONS: In conclusion, OMVs induce inflammation, fibrosis and suppression of albumin production, affecting the pathogenesis of cirrhosis. We believe that our study paves the way for the future prevention and treatment of cirrhosis.

Navitoclax improves acute-on-chronic liver failure by eliminating senescent cells in mice.

AIM: Acute-on-chronic liver failure (ACLF), a disease with poor prognosis, is reportedly caused by cellular senescence due to mitochondrial dysfunction. In this study, we described and analyzed the underlying mechanism of a novel approach for ACLF using ABT263/navitoclax (Navi) that selectively eliminates senescent cells. METHODS: Irradiation-induced senescent hepatocytes were used for in vitro evaluation of the effects of Navi on ACLF (n = 6 for each group). Lipopolysaccharide- and carbon tetrachloride-induced ACLF mouse model was used for in vivo evaluation of the effects of Navi administration compared with the control using one-way or two-way analysis of variance, followed by Student's t-test or Kruskal-Wallis test. The effects on the senescence-associated secretory phenotype (n = 8 for each group) and mitochondrial functions, including adenosine triphosphate concentration and membrane potential (n = 8 for each group), were investigated using real-time polymerase chain reaction, immunohistochemistry, and enzyme analysis. RESULTS: Navi eliminated irradiation-induced senescent hepatocytes in vitro, leading to non-senescent hepatocyte proliferation. Navi eliminated senescent cells in the liver in vivo, resulting in downregulation of mRNA expression of senescence-associated secretory phenotype factors, a decrease of liver enzymes, and upregulated proliferation of non-senescent cells in the liver. Regarding mitochondrial functional assessment in the liver, adenosine triphosphate concentration and membrane potential were upregulated after Navi administration in vitro and in vivo. CONCLUSIONS: Navi may ameliorate ACLF damage by eliminating senescent cells in the liver, downregulating senescence-associated secretory phenotype factors, and upregulating mitochondrial functions. We believe that this novel approach using Navi will pave the way for ACLF treatment.

Analysis of drug-induced liver-related adverse event trend reporting between 1997 and 2019.

AIM: This study aimed to analyze the current trends of drug-induced liver-related adverse events in the Food and Drug Administration Adverse Event Reporting System (FAERS) and Japanese Adverse Drug Event Report (JADER) databases. METHODS: The characteristics of implicated drugs were investigated by analyzing big data on drug-induced liver-related adverse events over the past 20 years in FAERS, comparing drug rankings between the JADER and FAERS databases, and calculating rankings of drugs inducing liver-related adverse events using the Medical Dictionary for Regulatory Activities Terminology. RESULTS: In the 452 272 cases registered in FAERS from 1997 to 2019, warfarin, paracetamol, and adalimumab were the drugs most related to drug-induced liver injury (DILI). In the 38 919 cases registered in JADER from 2004 to 2019, sorafenib, nivolumab, and herbal extracts were the drugs most related to DILI. No associations were found between the top 30 drugs in either of the two databases. Notably, the number of drug-induced liver-related adverse event reports and total adverse events has sharply increased in recent years. CONCLUSIONS: Although liver-related adverse events are largely caused by host immunity and other constitutional factors, differences in primary diseases, countries, and historical backgrounds lead to differences in the number of reports. Securing an appropriate database and a mechanism to collect real-time information on the frequency of adverse drug reactions is warranted.

Two rare pancreatic parenchymal hemorrhagic lesions associated with acute pancreatitis in acute liver failure: a case report and literature review.

Acute pancreatitis is an uncommon occurrence in acute liver failure. Furthermore, such cases are rarely complicated by parenchymal hemorrhages. Herein, we report the case of a 69-year-old male patient with multiple pancreatic parenchymal hemorrhages concomitant with acute liver failure. The patient underwent conservative treatment for acute liver failure caused by hepatitis B virus infection. Plain computed tomography on the 30th day revealed two high-density mass lesions in the pancreatic body and tail, which were suspected to be multiple pancreatic parenchymal hemorrhages. Despite restarting gabexate mesylate, the patient died of multiple organ failure on the 49th day. The clinical information of the present case and our literature review of 61 similar cases in 43 case reports identified via a systematic keyword search of the PubMed database, which described acute pancreatitis concomitant with acute hepatitis and acute liver failure, will aid physicians in the diagnosis and treatment of this rare condition.

Prognostic study of acute-on-chronic liver failure patients: Usefulness of the fibrosis-4 index.

Acute-on-chronic liver failure (ACLF) is a syndrome characterized by an acute deterioration of liver function in cirrhotic patients. Since treatment of this condition is difficult, its prevention is of paramount importance. The predictors of ACLF are yet to be identified. To determine the prognosis of cirrhotic and ACLF patients, we conducted a retrospective study to analyze each parameter in ACLF patients. Cirrhotic patients with serum total-bilirubin level ≥5.0 mg/dL and prothrombin time (PT) value ≤40% after acute insults were diagnosed with ACLF, whereas patients who met one of the above criteria were diagnosed with extended type of ACLF (EX-ACLF). Overall, in this study, 18 ACLF and 16 EX-ACLF patients retrospectively investigated between 2008 and 2020, and each data was analyzed during and before acute insults. In the analysis during acute insults, renal and coagulation functions showed significant differences between the ACLF and EX-ACLF groups. Furthermore, the mortality rate in the ACLF group was higher than that in the EX-ACLF group. In the analysis before acute insults, aspartate aminotransferase (AST), Fibrosis-4 (FIB-4) index score, and AST to platelet ratio index (APRI) showed significant differences between the two groups. Among these, the FIB-4 index score correlated best with ACLF severity for identifying cirrhotic patients with poor prognosis. The FIB-4 index is the most useful predictor of ACLF severity. Careful management of cirrhotic patients with a high FIB-4 index score is considered beneficial to prevent ACLF occurrence.

Small extracellular vesicles and liver diseases: From diagnosis to therapy.

Extracellular vesicles (EVs), especially small EVs (sEVs) derived from liver cells, have been the focus of much attention in the normal physiology and pathogenesis of various diseases affecting the liver. sEVs are approximately 100 nm in size, enclosed within lipid bilayers, and are very stable. The lipids, proteins, and nucleic acids, including miRNAs, contained within these vesicles are known to play important roles in intercellular communication. This mini-review summarizes the application of sEVs. First, liver diseases and the related diagnostic markers are described, and the current active status of miRNA research in diagnosis of hepatocellular carcinoma (HCC) is reported. Second, the biodistribution and pharmacokinetics of sEVs are described, and the liver is highlighted as the organ with the highest accumulation of sEVs. Third, the relationship between sEVs and the pathogenesis of liver disorders is described with emphesis on the current active status of miRNA research in HCC recurrence and survival. Finally, the possibility of future therapy using sEVs from mesenchymal stem (stromal) cells for cirrhosis and other diseases is described.

Synthesized HMGB1 peptide prevents the progression of inflammation, steatosis, fibrosis, and tumor occurrence in a non-alcoholic steatohepatitis mouse model.

AIM: Non-alcoholic steatohepatitis (NASH) with fibrosis eventually leads to cirrhosis and hepatocellular carcinoma. Thus, the development of therapies other than dietary restriction and exercise, particularly those that suppress steatosis and fibrosis of the liver and have a long-term beneficial effect, is necessary. We aimed to evaluate the therapeutic effects of the HMGB1 peptide synthesized from box A using the melanocortin-4 receptor-deficient (Mc4r-KO) NASH model mouse. METHODS: We performed short- and long-term administration of this peptide and evaluated the effects on steatosis, fibrosis, and carcinogenesis using Mc4r-KO mice. We also analyzed the direct effect of this peptide on macrophages and hepatic stellate cells in vitro and performed lipidomics and metabolomics techniques to evaluate the effect. RESULTS: Although this peptide did not show direct effects on macrophages and hepatic stellate cells in vitro, in the short-term administration model, we could confirm the reduction of liver damage, steatosis, and fibrosis progression. The results of lipidomics and metabolomics suggested that the peptide might ameliorate NASH by promoting lipolysis via the activation of fatty acid ß-oxidation and improving insulin resistance. In the long-term administration model, this peptide prevented progression to cirrhosis but retained the steatosis state, that is, the peptide prevents the progression to "burnt-out NASH." This peptide inhibited carcinogenesis by about one-third. CONCLUSION: This HMGB1 peptide can reduce liver damage, improve fibrosis and steatosis, and inhibit carcinogenesis, suggesting that the peptide would be a new treatment candidate for NASH and can contribute to the long-term prognosis for patients with NASH.

Cumulative risk of developing a new symptom in patients with primary biliary cholangitis and its impact on prognosis.

Background and Aim: Symptoms of primary biliary cholangitis (PBC) frequently impair one's quality of life (QOL). Nonetheless, with improved treatment, the prognosis of PBC also improves. QOL plays an important role in patients with PBC. In this study, we aimed to reevaluate the transition of new symptom development in PBC and its predictive factors. Methods: This retrospective multicenter study enrolled 382 patients with PBC for symptom analysis. The impact of a newly developed symptom on PBC prognosis was investigated by Kaplan-Meier analysis with propensity score matching and logistic progression analysis. Results: The cumulative risk of developing a new symptom after 10 and 20 years of follow-up was 7.6 and 28.2%, and specifically that of pruritus, which was the most common symptom, was 6.7 and 23.3%, respectively. In Cox hazard risk analysis, serum Alb level (hazard ratio [HR], 1.097; 95% confidence interval [CI], 1.033-1.165; P = 0.002), the serum D-Bil level (HR, 6.262; 95% CI, 2.522-15.553, P < 0.001), and Paris II criteria (HR, 0.435; 95% CI, 0.183-1.036; P = 0.037) were significant independent predictors of a new symptom. Kaplan-Meier analysis showed that the overall survival and liver-related death were not significant between patients with and without a new symptom. Conclusion: The cumulative risk of new symptom development is roughly 30% 20 years after diagnosis and could be predicted by factors including serum albumin levels, serum D-Bil level, and Paris II criteria.

Basic points to consider regarding the preparation of extracellular vesicles and their clinical applications in Japan.

In recent years, extracellular vesicles (EVs) have attracted attention as a new therapeutic tool. In Europe, the United States, and Asia, there is an accelerating trend of moving beyond basic research on clinical trials. However, treatment using EVs is still in the research and development stage, and the general public has insufficient awareness and understanding of the risks involved in ensuring safety and efficacy, the status of laws and regulations, and global research and development trends regarding their use. The Japanese Society for Regenerative Medicine, which has promoted the research and development of regenerative medicine, an innovative medical technology based on the principle of delivering it safely, effectively, and promptly, including the establishment of laws and regulations, would like to express two positions in light of the rapid development of therapies using EVs: 1) concern about treatments that are based solely on the discretion of medical practitioners, and 2) active promotion of treatments based on sound scientific evidence. Because EVs are released from cells, there are many similarities between EVs and processed cells in terms of manufacturing processes and safety hazards. As for efficacy, the mechanism of action of EVs is still unclear, as is the case with specified processed cellsb; in such cases, it is difficult to measure potency, identify efficacy-related quality attributes, and evaluate the comparability of quality before and after a change in the manufacturing process. In other words, the number of quality attributes that can be obtained for EVs is limited because of their complex characteristics, and it is difficult to grasp their quality through specifications and characterization. Therefore, while designing a quality control strategy for EVs, it is important to ensure the quality of the final product (EVs) by controlling the raw materials and manufacturing process. On the contrary, since EVs do not contain living cell components and are not classified into specified processed cells, non-commercial clinical research on treatments using EVs and individual medical treatments with EVs at the discretion of medical practitioners are out of the scope of the Act on the Safety of Regenerative Medicine of Japan. At present, there are no relevant laws or regulations for the use of EVs other than the Medical Practitioners' Act and the Medical Care Act in Japan. Therefore, there is a concern that treatment will be performed without sufficient objective evaluation of the scientific basis for safety and efficacy. Despite these concerns, the development of therapies using EVs is underway worldwide. This could potentially lead to a wide variety of new therapeutic areas if the methods needed to stably secure and mass cultivate cells as raw materials and the technologies needed for the mass production of EVs can be developed, in addition to understanding the risks involved and developing relevant laws and regulations. As part of the Japanese Society for Regenerative Medicine, we will continue to work on the development of these methods and technologies and hope that such a promising field will be promoted with a high level of safety before reaching the public.

Utility of autologous fecal microbiota transplantation and elucidation of microbiota in diversion colitis.

Objectives: Diversion colitis (DC) is an inflammatory disorder caused by interruption of the fecal stream and subsequent nutrient deficiency from luminal bacteria. The utility of fecal microbiota transplantation (FMT) for DC was recently investigated; however, the precise pathogenesis of this condition remains unclear. This study aimed to evaluate the utility of autologous FMT in DC and to determine the related changes in the intestinal microbiota. Methods: Autologous FMT was performed to reestablish the intestinal microbiota in five patients (average age, 64.6 ± 8.3 years) with DC. They underwent double-ended colostomy. We assessed the diverted colon by endoscopy and evaluated the microbiota before and after FMT using the 16S rRNA gene sequencing method. Results: All five patients had mild inflammation (ulcerative colitis endoscopic index of severity [UCEIS] 2-3) in the diverted colon based on the colonoscopic findings. Three patients presented with symptoms, such as tenesmus, mucoid stool, and bloody stool. With FMT treatment, all patients achieved endoscopic remission (UCEIS score of 0 or 1) and symptomatic improvement. We observed a significantly decreased α-diversity in DC patients compared to healthy controls. The frequency of aerobic bacteria, such as Enterobacteriaceae, in the diverted colon decreased after autologous FMT. Conclusions: This study was the first to show that the microbiota in the diverted colon was significantly affected by autologous FMT. Since interruption of the fecal stream is central to the development of DC, FMT can be considered a promising treatment.

Hydrogen-producing small intestinal bacterial overgrowth is associated with hepatic encephalopathy and liver function.

Overt hepatic encephalopathy (HE) is one of the complications of liver cirrhosis (LC), which negatively affects the prognosis and quality of life of patients. Small intestinal bacterial overgrowth (SIBO) is significantly associated with LC and its complications, including HE. We investigated the relationship between SIBO and LC, and the difference between hydrogen-producing and methane-producing SIBO (H-SIBO and M-SIBO, respectively). This is a prospective cohort study of 107 cases. Breath measurements of hydrogen and methane concentrations were performed for the diagnosis of SIBO. The study cohort included 81 males with a median age of 70 (40-86) years, and SIBO was detected in 31 cases (29.0%). There were no significant differences between the SIBO positive and SIBO negative groups. Reclassification into H-SIBO (16 cases) and others (91 cases) was performed, and the Child-Pugh score was only derived in the multivariate logistic analysis (P = 0.028, odds ratio 1.39, 95% confidence interval 1.04-1.85). Furthermore, H-SIBO was significantly associated with covert HE in chi-square test (50.0% vs. 24.2%, P = 0.034). In addition, we evaluated the therapeutic response on SIBO of rifaximin in eight covert HE patients. 20% patients with M-SIBO and 67% patients with H-SIBO showed an improvement of the breath test. In conclusion, H-SIBO, but not M-SIBO, is significantly associated with liver function, and rifaximin might be more effective for covert HE with H-SIBO. Therefore, the diagnosis of SIBO, including the classification as H-SIBO and M-SIBO, might help to determine the choice of treatment for HE.

A novel prostaglandin I2 agonist, ONO-1301, attenuates liver inflammation and suppresses fibrosis in non-alcoholic steatohepatitis model mice.

BACKGROUND: ONO-1301 is a novel long-lasting prostaglandin (PG) I2 mimetic with inhibitory activity on thromboxane (TX) A2 synthase. This drug can also induce endogenous prostaglandin (PG)I2 and PGE2 levels. Furthermore, ONO-1301 acts as a cytokine inducer and can initiate tissue repair in a variety of diseases, such as pulmonary hypertension, pulmonary fibrosis, cardiac infarction, and obstructive nephropathy. In this study, our aim was to evaluate the effect of ONO-1301 on liver inflammation and fibrosis in a mouse model of non-alcoholic steatohepatitis (NASH). METHODS: The therapeutic effects of ONO-1301 against liver damage, fibrosis, and occurrence of liver tumors were evaluated using melanocortin 4 receptor-deficient (Mc4r-KO) NASH model mice. The effects of ONO-1301 against macrophages, hepatic stellate cells, and endothelial cells were also evaluated in vitro. RESULTS: ONO-1301 ameliorated liver damage and fibrosis progression, was effective regardless of NASH status, and suppressed the occurrence of liver tumors in Mc4r-KO NASH model mice. In the in vitro study, ONO-1301 suppressed LPS-induced inflammatory responses in cultured macrophages, suppressed hepatic stellate cell (HSC) activation, upregulated vascular endothelial growth factor (VEGF) expression in HSCs, and upregulated hepatocyte growth factor (HGF) and VEGF expression in endothelial cells. CONCLUSIONS: The results of our study highlight the potential of ONO-1301 to reverse the progression and prevent the occurrence of liver tumors in NASH using in vivo and in vitro models. ONO-1301 is a multidirectional drug that can play a key role in various pathways and can be further analyzed for use as a new drug candidate against NASH.

Longitudinal increase in albumin-bilirubin score is associated with non-malignancy-related mortality and quality of life in patients with liver cirrhosis.

Due to the developments in the treatment for hepatitis, it is possible to prevent the progression of liver fibrosis and improve patients' prognosis even if it has already led to liver cirrhosis (LC). Consequently, a two-step study was conducted. To begin with, a retrospective study was conducted to identify the potential predictors of non-malignancy-related mortality from LC. Then, we prospectively analyzed the validity of these parameters as well as their association with patients' quality of life. In the retrospective study, 89 cases were included, and the multivariate Cox regression analysis indicated that age (P = 0.012), model for end-stage liver disease (MELD) score (P = 0.012), and annual rate of change of the albumin-bilirubin (ALBI) score (P < 0.001) were significantly associated with LC prognosis. In the prospective study, 70 patients were included, and the patients were divided into cirrhosis progression and non-progression groups. The univariate logistic regression analysis indicated the serum procollagen type III N-terminal peptide level (P = 0.040) and MELD score (P = 0.010) were significantly associated with the annual rate of change of the ALBI score. Furthermore, the mean Chronic Liver Disease Questionnaire score worsened from 5.3 to 4.9 in the cirrhosis progression group (P = 0.034). In conclusion, a longitudinal increase in the ALBI score is closely associated with non-malignancy-related mortality and quality of life.

Use of a Deep Learning Approach for the Sensitive Prediction of Hepatitis B Surface Antigen Levels in Inactive Carrier Patients.

Deep learning is a subset of machine learning that can be employed to accurately predict biological transitions. Eliminating hepatitis B surface antigens (HBsAgs) is the final therapeutic endpoint for chronic hepatitis B. Reliable predictors of the disappearance or reduction in HBsAg levels have not been established. Accurate predictions are vital to successful treatment, and corresponding efforts are ongoing worldwide. Therefore, this study aimed to identify an optimal deep learning model to predict the changes in HBsAg levels in daily clinical practice for inactive carrier patients. We identified patients whose HBsAg levels were evaluated over 10 years. The results of routine liver biochemical function tests, including serum HBsAg levels for 1, 2, 5, and 10 years, and biometric information were obtained. Data of 90 patients were included for adaptive training. The predictive models were built based on algorithms set up by SONY Neural Network Console, and their accuracy was compared using statistical analysis. Multiple regression analysis revealed a mean absolute percentage error of 58%, and deep learning revealed a mean absolute percentage error of 15%; thus, deep learning is an accurate predictive discriminant tool. This study demonstrated the potential of deep learning algorithms to predict clinical outcomes.

Novel Strategy for Diagnosis of Focal Nodular Hyperplasia Using Gadolinium Ethoxybenzyl Diethylenetriaminepentaacetic Acid: Enhanced Magnetic Resonance Imaging and Magnetic Resonance Elastography.

Focal nodular hyperplasia (FNH) is the second most frequent benign liver tumor, and it is a fiber-rich stiff lesion. Typically, FNH can be diagnosed by imaging without biopsy. However, liver biopsy and diagnostic resection may be required to differentiate atypical FNH from other liver tumors, such as hepatocellular adenoma (HCA). Therefore, improved noninvasive diagnostic methods are needed. We experienced 2 cases where combination of magnetic resonance elastography (MRE) and gadolinium ethoxybenzyl diethylenetriaminepentaacetic acid (Gd-EOB-DTPA)-enhanced magnetic resonance imaging (MRI) helped diagnose FNH. A 36-year-old woman and 17-year-old boy with liver tumors measuring 40 mm in diameter each showed hypointense nodule centers, indicating a central scar, surrounded by hyperintense signals during the hepatobiliary phase of Gd-EOB-DTPA-enhanced MRI. To rule out HCA, we performed MRE and liver biopsy. On MRE, the mean stiffness of the mass was 11.6 kPa (mean stiffness of the background liver was 1.7 kPa) and 11.1 kPa (mean stiffness of the background liver was 2.4 kPa) in the first and second patients, respectively. Histological examination of both specimens showed CK7-positive bile-ductular proliferations, abundant fibrous tissue, and few Ki-67-positive cells. Based on these results, we diagnosed these tumors as FNH. Combination of Gd-EOB-DTPA-enhanced MRI and MRE can evaluate the character and stiffness of lesion and help in the diagnosis of FNH.